Exposure and Toxicity Factors in Health Risk Assessment of Heavy Metal(loid)s in Water

Abstract

Heavy metal(loid) (HM) contamination in water arises from various anthropogenic activities and natural processes, posing risks to human health through ingestion and dermal absorption. Although numerous studies have assessed health risks associated with HMs in water, inconsistencies in the selection of exposure and toxicity factors limit comparability and reliability across studies. To address this gap, the aim of this review was to provide a comprehensive synthesis of exposure and toxicity factors used in health risk assessment (HRA) of HMs in water. The objectives were to evaluate the variability in ingestion, body weight, exposure duration and frequency, and dermal contact parameters, as well as in reference doses and cancer slope factors and to propose standardized values and statistical distributions for more consistent risk estimation. A systematic search of the Scopus database retrieved 806 studies, from which highly cited articles (≥100 citations) and recent publications (2023–2025) were prioritized for analysis. The findings revealed substantial variability in factors and showed that probabilistic approaches, particularly Monte Carlo simulation, were increasingly applied and provided more reliable estimates than traditional deterministic methods. The highest agreement was observed for exposure frequency for ingestion (365 days/year) and skin surface area (18,000 cm²), each applied in 75.5% of cases. By identifying inconsistencies in current practices and proposing standardized exposure and toxicity values and distributions for water, this review is expected to offer practical recommendations to improve the robustness, reliability, and comparability of HRAs, ultimately informing more effective policy-making and water management practices.

1. Introduction

Water is a fundamental natural resource, crucial for human health, daily life, and overall socio-economic development [1,2,3]. However, water quality is increasingly threatened by contamination from various pollutants, among which heavy metal(loid)s (HMs) such as As, Cd, Cr, Pb, Zn, Cu, Ni, Mn, Fe, Mo, Hg, Sn, Ba, Sr, Se, Co, V, Sb, and Al have been the focus of numerous studies evaluating their potential impacts on human health [4,5,6]. These elements are characterized by their toxicity, persistence in the environment, and bioaccumulation in organisms [7,8].

Heavy metal(loid)s can enter water bodies through both natural and anthropogenic pathways, as shown in Figure 1. Rapid economic development, particularly in industry, mining, and agricultural activities, has made anthropogenic inputs the dominant source of HMs in water. When it comes to natural sources, HMs are naturally present in the Earth’s crust, most often in the form of sulfides and oxides [9]. Through the interaction between water and rocks, they are released into aquatic systems. However, beyond these, a fraction of HMs in soils and sediments occurs as exchangeable cations or adsorbed onto mineral surfaces [10]. These forms are more readily mobilized and can enter groundwater. In particular, HM ions frequently associate with Fe/Mn oxides or carbonates. Although such sorption can provide temporary immobilization, changes in pH or redox conditions may trigger their remobilization into groundwater [11]. In addition, volcanic eruptions and wildfires release HMs into the environment, from where they percolate through soil and eventually leaching into surface and groundwater [12]. As for anthropogenic sources, they include wastewater discharge from industries such as chemical production and electroplating, mining and metallurgical activities, municipal wastewater effluents, leachate from landfills, excessive use of artificial fertilizers in agriculture, as well as fuel combustion in traffic and industrial facilities [13,14,15,16].

Long-term exposure to HMs, even at low concentrations, has been associated with a wide range of adverse health effects, illustrated in Figure 1. As shown, HM toxicity can affect multiple organ systems, manifesting as lung, kidney, and liver cancers, pneumonia, hypertension, cardiovascular disease, infertility, osteoporosis, and cognitive impairment [17]. Consequently, assessing the health risks associated with HMs in water is of great importance for environmental monitoring, risk management, and the protection of human health.

The health risk assessment (HRA) provides a structured methodology to evaluate the potential adverse health outcomes from exposure to hazardous substances in the environment. According to the United States Environmental Protection Agency (USEPA), the HRA framework consists of four key steps: hazard identification, dose–response assessment, exposure assessment, and risk characterization [18]. Within this framework, exposure assessment connects measured concentrations of contaminants in water with the actual intake by humans. The accuracy of this step directly influences the reliability of the final risk estimates. Exposure assessment relies on a set of parameters collectively referred to as exposure factors, which describe the frequency, duration, and time of human contact with contaminants [19]. These include ingestion rate (IR), exposure frequency (EF), exposure duration (ED), body weight (BW), averaging time (AT), skin surface area (SA), exposure time (ET), dermal permeability coefficient (Kp), and conversion factor (CF). Each of these parameters plays an essential role in calculating chronic daily intake (CDI), which is subsequently used to quantify both non-carcinogenic and carcinogenic risks. Importantly, exposure factors differ across population groups (adults versus children) and across exposure pathways. For water, ingestion is the primary pathway of concern, while dermal contact is also relevant, particularly in scenarios involving bathing or occupational exposure. In addition to exposure factors, toxicity values, such as the reference dose (RfD) and cancer slope factor (SF), are equally critical, as non-carcinogenic and carcinogenic risks directly depend on them.

Variability and uncertainty in exposure and toxicity factors remain a major challenge in health risk assessments of HMs. Researchers frequently adopt different values for the same parameters [20,21,22], and in many cases these differ from the USEPA recommendations, which may lead to inconsistent or incomparable risk estimates. This situation highlights the need for a systematic review and comparison of commonly applied values against standardized references. Accordingly, the aim of our study is to identify common practices, examine existing gaps and inconsistencies, and propose recommendations that will strengthen the consistency and scientific basis of future risk assessments. Therefore, this study compiled and critically evaluated the exposure and toxicity factors most frequently employed in health risk assessments of 19 HMs in water (As, Cd, Cr, Pb, Zn, Cu, Ni, Mn, Fe, Mo, Hg, Sn, Ba, Sr, Se, Co, V, Sb, and Al), compared them with reference values from the USEPA, and highlighted their implications for both deterministic and probabilistic risk assessment approaches.

2. Materials and Methods

2.1. Data Extraction and Categorization

This study conducted a systematic review on HMs in water using the Scopus database search engine, carried out in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) methodology. Scopus was preferred as it is the world’s largest abstract and citation database, encompassing a broader spectrum of journals than many alternative sources. The use of the PRISMA methodology ensures that the review is thorough and systematic, while minimizing potential errors.

In this review, the following research questions were addressed: (i) which values of exposure and toxicity parameters have been applied in the published literature on HRA of HMs in water; (ii) how these reported values align with USEPA recommendations; (iii) what level of variability exists across studies, including differences between adult and child populations; (iv) to what extent deterministic and probabilistic (Monte Carlo) approaches are applied; and (v) what recommendations can be drawn to support more consistent and scientifically robust HRAs of HMs in water. To answer these questions, in the first step, a comprehensive Scopus search was performed to capture the general trends in the field. Workflow is illustrated in Figure S1.

The query combined keywords related to aquatic environments (e.g., water, groundwater, river), heavy metals (e.g., heavy metal(loid)s, potentially toxic elements), exposure parameters (e.g., ingestion rate, body weight, exposure frequency), and health risk assessment terms (e.g., health risk, hazard index, cancer risk), as well as probabilistic approaches (e.g., probabilistic, Monte Carlo). The search was limited to peer-reviewed articles published in English. The full Scopus query was as follows: (TITLE (water OR groundwater OR river OR lake OR lagoon OR well OR pond OR stream OR creek) AND TITLE (“heavy metal(loid)s” OR “potentially toxic elements” OR “heavy metals” OR “toxic elements” OR “trace metals” OR “trace elements” OR “metal content”) AND TITLE-ABS-KEY (“ingestion rate” OR “body weight” OR “exposure frequency” OR RfD OR “slope factor” OR adults OR children) AND TITLE-ABS-KEY (“health risk” OR “hazard index” OR “cancer risk” OR ILCR OR ELCR OR “health hazard” OR TCR) OR TITLE-ABS-KEY (probabilistic OR “Monte Carlo”)) AND (LIMIT-TO (DOCTYPE, “ar”)) AND (LIMIT-TO (LANGUAGE, “English”)). In the first comprehensive search, a total of 806 articles were retrieved.

The trend of the number of publications and citations is presented in Figure 2. The first article in this field was recorded in 1994, and up to around 2016 the annual number of publications did not exceed 10. A sharp expansion of the field has been observed since 2016, with both the number of articles and citations increasing year by year, showing an average annual growth rate of about 44% and 46%, respectively. If this average growth is projected to continue, the year 2025 is expected to record 240 articles and 8459 citations.

In this study, a keyword analysis was also conducted using the VOS (visualization of similarities) viewer software v. 1.6.20 (Centre for Science and Technology Studies, Leiden University, The Netherlands). Data were exported from Scopus in CSV format, including full records and cited references. A co-occurrence keyword analysis was applied, where the unit of analysis was defined as all keywords, and the full counting method was used. A total of 5279 keywords were identified across 806 retrieved articles. For further analysis, a minimum occurrence threshold of 10 was applied, resulting in 373 keywords meeting the criterion. The co-occurrence network visualization of all keywords is shown in Figure 3. In this keyword network visualization each circle represents a keyword. The circle size reflects its frequency of occurrence, the distance between circles indicates the strength of their co-occurrence, and the color denotes the cluster to which each keyword belongs. The keyword most frequently observed was “risk assessment” (558 occurrences), followed by “heavy metal” (497), “health risk” (457), “heavy metals” (388), “human” (375), and “environmental monitoring” (314). With regard to individual HMs, the most frequent keywords were “lead” (270), “cadmium” (263), “chromium” (219), “copper” (213), and “zinc” (209). Concerning the investigated aquatic media, the keyword “rivers” occurred most often (189), followed by “groundwater” (188) and “drinking water” (175). It was observed that keywords such as “monte carlo method”, “monte carlo analysis”, and “monte carlo simulation” started to appear only from 2022 to 2023. This finding is further supported by the VOSviewer overlay visualization of all keywords by average publication year (Figure S2), where can be seen that Monte Carlo emerges in the literature around 2023. This indicates that the application of Monte Carlo simulation and, more generally, probabilistic risk assessment has only recently begun to be used in the water risk assessment.

Additionally, keywords occurring 10 or more times were organized into four clusters: pink, cyan, brown, and green (Figure 3). The pink cluster referred to health risk assessment of HMs, including the determination of water pollution levels and comparison with regulations such as those of the World Health Organization. In the cyan cluster, the presence of keywords related to soil or sediment was observed, indicating that this cluster refers to soil/sediment-to-water pollution. The brown cluster included keywords related to the bioavailability of various HMs, addressing how HMs in water are transferred to humans, animals, or plants. Finally, the green cluster was characterized by keywords of individual HMs and may be associated with their detection in different water bodies.

After the comprehensive search (806 records), a targeted query was applied to focus the review on recent and domain-relevant literature. The refined search was limited to English-language articles published between 2023 and 2025, with irrelevant subject areas excluded (e.g., social sciences, materials science, physics, immunology, health professions, veterinary, nursing, neuroscience, mathematics, economics, business, biochemistry, pharmacology, energy, computer science, and decision science). This refinement yielded 306 articles for screening. In addition, to capture established, widely referenced articles, a high-impact subset was defined from the 806 studies retrieved in the search, as those with ≥100 citations, yielding 49 articles. This subset was identified by sorting within the Scopus interface. The timeframe of 2023–2025 was chosen to emphasize current research trends, while the inclusion of highly cited earlier works ensured that influential and foundational contributions prior to 2023 were also represented.

Two datasets were combined (n = 355), one comprising highly cited papers and the other limited to the years 2023–2025. Titles and abstracts were first screened, followed by full-text review and extraction of exposure parameter values. During the screening of titles and abstracts, 145 studies were excluded because they did not address health risk assessment of HMs in water, but focused instead on other media or analytes. This left 210 articles for full-text review, of which one could not be accessed. Among the remaining studies, 101 were excluded either because parameter values were not reported or the applied formulas were inconsistent or deviated from USEPA recommendations. Finally, 108 studies met the criteria and were included into the review. The PRISMA flowchart, showing the number of studies at each stage, is presented in Figure 4.

In order to visualize the global distribution of the extracted studies, the number of studies by country has been mapped, as shown in Figure S3. It was observed that this field has been most extensively investigated in the southeastern part of Asia, particularly in India, China, and Bangladesh. In addition to these countries, HMs in groundwater have also been sporadically studied in some African countries, such as Nigeria, Algeria, and Ethiopia, as well as in certain European countries, including Serbia, Romania, Poland, and Italy. This uneven distribution of studies highlights a geographical research gap, as large regions such as Europe, North America, and Australia remain underrepresented, limiting the global perspective on health risks from HMs in water.

2.2. Health Risk Methodology

The health risk assessment of HMs in water is conducted following the methodology developed by the USEPA [18]. The methodology is shown in Figure 5 and consists of four steps. First, it is necessary to identify the harmful HM substances, whether they have the potential to cause adverse health effects in humans and under which conditions, as well as whether these are non-carcinogenic or carcinogenic effects. Then follows the dose–response assessment, in which the relationship between the concentration of the toxic substance and adverse health effects is defined. In this way, parameters such as the reference dose (RfD) and the cancer slope factor (SF) are obtained. In the third step, exposure assessment, the exposed population is identified, most often divided into adults and children. In addition, the exposure pathways to toxic substances are identified. For water, these are ingestion, which is dominant, and dermal contact. Then, taking into account exposure parameters such as exposure duration or exposure frequency, the intake of each toxic element per kilogram of body weight per day is determined, which is referred to as chronic daily intake (CDI). Finally, risk characterization follows, with the aim of quantifying both non-carcinogenic and carcinogenic risk, which can then be compared with permitted thresholds.

In the methodology of health risk assessment, quantification of risks allows the results to be compared with regulatory thresholds, thereby clearly determining whether there is a risk to human health, which is essential for making decisions on health protection and remediation. As already mentioned, the risk can be non-carcinogenic or carcinogenic. The calculation of these two risks begins with the estimation of chronic daily intake for both exposure pathways according to the following Equations (1) and (2).

$${\mathrm{C}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{i}\mathrm{n}\mathrm{g}}={\displaystyle \frac{{\mathrm{C}}_{\mathrm{i}}\times \mathrm{I}\mathrm{R}\times \mathrm{E}\mathrm{F}\times \mathrm{E}\mathrm{D}}{\mathrm{B}\mathrm{W}\times \mathrm{A}\mathrm{T}}}$$

(1)

$${\mathrm{C}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{m}}={\displaystyle \frac{{\mathrm{C}}_{\mathrm{i}}\times \mathrm{S}\mathrm{A}\times {\mathrm{K}\mathrm{p}}_{\mathrm{i}}\times \mathrm{E}\mathrm{T}\times \mathrm{E}\mathrm{F}\times \mathrm{E}\mathrm{D}\times \mathrm{C}\mathrm{F}}{\mathrm{B}\mathrm{W}\times \mathrm{A}\mathrm{T}}}$$

(2)

where CDI_i,ing and CDI_i,derm represent the chronic daily intake through ingestion and dermal exposure, respectively, in mg/(kg×day) for heavy metal i. The other parameters appearing in Equations (1) and (2) are referred to as exposure parameters: C_i (mg/L) is the concentration of HM i in water, IR (L/day) is the ingestion rate, EF (days/year) is the exposure frequency, ED (years) is the exposure duration, BW (kg) is the body weight, AT (days) is the averaging time, SA (cm²) is the skin surface area, Kp_i (cm/h) is the dermal permeability coefficient of HM i, ET (h/day) is the exposure time, and CF (L/cm³) is the unit conversion factor. All of these parameters will be explained in the following section.

Once the CDI is calculated, the next step is the estimation of risk, namely the non-carcinogenic risk quantified by the hazard index (HI) and the carcinogenic risk quantified by the incremental lifetime cancer risk (ILCR), using Equations (3)–(10).

$${\mathrm{H}\mathrm{Q}}_{\mathrm{i},\mathrm{i}\mathrm{n}\mathrm{g}}={\displaystyle \frac{{\mathrm{C}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{i}\mathrm{n}\mathrm{g}}}{{\mathrm{R}\mathrm{f}\mathrm{D}}_{\mathrm{i},\mathrm{i}\mathrm{n}\mathrm{g}}}}$$

(3)

$${\mathrm{H}\mathrm{Q}}_{\mathrm{i},\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{m}}={\displaystyle \frac{{\mathrm{C}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{m}}}{{\mathrm{R}\mathrm{f}\mathrm{D}}_{\mathrm{i},\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{m}}}}$$

(4)

$${\mathrm{H}\mathrm{Q}}_{\mathrm{i}}={\mathrm{H}\mathrm{Q}}_{\mathrm{i},\mathrm{i}\mathrm{n}\mathrm{g}}+{\mathrm{H}\mathrm{Q}}_{\mathrm{i},\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{m}}$$

(5)

$$\mathrm{H}\mathrm{I}={\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{H}\mathrm{Q}}_{\mathrm{i}}$$

(6)

$${\mathrm{C}\mathrm{R}}_{\mathrm{i},\mathrm{i}\mathrm{n}\mathrm{g}}={\mathrm{C}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{i}\mathrm{n}\mathrm{g}}\times {\mathrm{S}\mathrm{F}}_{\mathrm{i},\mathrm{i}\mathrm{n}\mathrm{g}}$$

(7)

$${\mathrm{C}\mathrm{R}}_{\mathrm{i},\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{m}}={\mathrm{C}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{m}}\times {\mathrm{S}\mathrm{F}}_{\mathrm{i},\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{m}}$$

(8)

$${\mathrm{C}\mathrm{R}}_{\mathrm{i}}={\mathrm{C}\mathrm{R}}_{\mathrm{i},\mathrm{i}\mathrm{n}\mathrm{g}}+{\mathrm{C}\mathrm{R}}_{\mathrm{i},\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{m}}$$

(9)

$$\mathrm{I}\mathrm{L}\mathrm{C}\mathrm{R}={\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{C}\mathrm{R}}_{\mathrm{i}}$$

(10)

The hazard index represents the sum of all hazard quotients (HQ), which are obtained by dividing CDI by the reference dose (RfD, mg/(kg × day)) of a given HM for both exposure pathways. When either HQ or HI exceeds 1, the non-carcinogenic risk is considered unacceptable. Otherwise, it is regarded as negligible. The incremental lifetime cancer risk represents the sum of all cancer risks (CR), which are obtained by multiplying CDI with the corresponding cancer slope factor (SF, (kg × day)/mg) of a given HM for each exposure pathway. An ILCR below 1 × 10⁻⁶ is considered a negligible risk, between 1 × 10⁻⁶ and 1 × 10⁻⁴ an acceptable risk, and above 1 × 10⁻⁴ an unacceptable carcinogenic risk. It is important to emphasize that HI, HQ, CR, and ILCR are dimensionless, while CDI is expressed in mg/(kg × day).

3. Exposure and Toxicity Factors

The exposure and toxicity parameters used in health risk assessment are discussed individually in the subsections below. It was observed that approximately half as many studies have addressed dermal risk associated with HMs present in water, with the majority focusing on ingestion exposure. Additionally, the studies were primarily focused on the assessment of non-carcinogenic risk, whereas carcinogenic risk was examined to a much lesser extent. Moreover, the extent to which individual elements have been examined varies considerably. Specifically, HMs such as As, Cd, Cr, Pb, Zn, Cu, Ni, Mn, and Fe have been the focus of numerous studies evaluating their potential impacts on human health. In contrast, elements such as Mo, Hg, Sn, Ba, Sr, Se, Co, V, Sb, and Al have been assessed in a substantially smaller number of such investigations. Tables S1 and S2 present the collected results of exposure and toxicity parameters from the reviewed studies, showing the frequency of occurrence of a certain value for each parameter.

Differences in the values of these parameters across studies can lead to substantially different health risk estimates. Even a slightly lower or higher value for a given parameter directly influences the calculated risks. For example, if an ingestion rate of 1 L is applied and compared with 1.5 L, while all other parameters remain constant, the resulting health risk (whether carcinogenic or non-carcinogenic) will differ by 50%. Such differences may affect whether the risk is classified as acceptable or unacceptable, and they also complicate comparisons across studies or the conduct of meta-analyses.

3.1. Ingestion Rate

The ingestion rate is explained as the amount of water an individual consumes through ingestion, expressed in L/day. In the majority of studies, the most commonly reported value for this parameter is 2 L/day for adults [4,23,24,25] and 1 L/day [3,26,27] for children. These values were employed in 41% and 31.3% of examined studies, respectively. However, regarding adults, values of 2.2 L/day [28,29] and 2.5 L/day [30,31,32] can also often be found in the literature, while for children, these values are 0.64 L/day [33,34,35] and 0.78 L/day [22,36]. Sporadically, significantly higher or lower values than those previously mentioned for adults were also used in the studies, such as 3 L/day [37], 3.5 L/day [38], 1.8 L/day [39], 1.23 L/day [40], and 1 L/day [41]. In the case of children, higher values such as 1.8 L/day [20,42], 1.5 L/day [43], and 2 L/day [44,45], or lower values such as 0.51 L/day [40] can also be found. Additionally, some authors have used the same ingestion rate values for both children and adults, with reported values of 1 L/day [46], 1.8 L/day [6], and 2 L/day [47].

3.2. Exposure Frequency

Exposure frequency is explained as the number of days in a year during which individuals are exposed to water consumption or dermal contact with it, expressed in days/year. It captures how often exposure occurs within a single year. For both adults and children, the same values of this parameter were used, i.e., there is no separation into population-specific values, nor was there a separation between dermal contact and ingestion. In the literature, authors used only two values: 350 days/year and 365 days/year. In studies that examined risks originating from both exposure routes, the dominant number of studies, 68% of them, used an exposure frequency value of 365 days/year [48,49,50,51], while the rest employed 350 days/year [52,53,54]. As for studies that examined only the risk from water ingestion, the percentage is somewhat higher in favor of an exposure frequency of 365 days/year [1,55,56,57], amounting to 83.3%, while the rest employed 350 days/year [27,58]. In summary, 75.5% of all studies reporting this factor used a values of 365 days/year. Interestingly, some authors [20,59,60] used different values of this parameter for the ingestion exposure route and the dermal exposure route, namely 365 days/year for the former and 350 days/year for the latter.

3.3. Exposure Duration

Exposure duration is the total time span over which the exposure occurs, illustrating how many years the exposure continues over a lifetime. Unlike exposure frequency, exposure duration differs for children and adults. In the case of adults, the largest number of studies (38.6%) used a value of 30 years [61,62,63,64], followed by 70 years (34.7%) [21,65,66]. Occasionally, values of 24 years [39], 40 years [67,68], and 64 years [55,69] also appeared. For children, there was greater consistency, with 6 years predominantly reported in the literature [6,66,70,71], appearing in 72.4% of studies. In addition to this exposure duration value, some authors used 4 years [40,68], 10 years [72,73], as well as 12 years [36,55].

3.4. Body Weight

In health risk assessment, body weight (kg) determines the dose of a contaminant per unit of body mass, meaning that individuals with lower body weight, such as children, receive a proportionally higher dose from the same level of intake. It is considered as the average body weight. In the existing literature, authors most frequently used a value of 70 kg for adults [67,74,75,76] and 15 kg [62,77,78,79] for children, which accounted for 54.7% and 57.4% of studies, respectively. However, this parameter varied widely, with the lowest value being 48.8 kg for adults [45] and 6 kg for children [6], while the highest values were 80 kg for adults [68] and 42.6 kg for children [52]. Regarding adults, in addition to the aforementioned values, authors have used body weights of 55 kg [61,80], 57.5 kg [55,81,82], 60 kg [35,36], 62.5 kg [83], 65 kg [41,84], and 72 kg [85]. For children, these were values of 10 kg [86], 16.3 kg [44,45], 20 kg [34], 25 kg [87], 30 kg [72,88], and 32.7 kg [89]. Additionally, it was observed that a large number of studies (40.6%) used a specific combination for body weights of adults and children, namely 70 kg for adults and 15 kg for children. Differences in body weight reported across studies often reflect regional variations in body size. Genetics, diet, lifestyle, and socio-economic factors all shape these differences, so it is not surprising that body weights used in the health risk assessment showed this much variation.

3.5. Averaging Time

Averaging time is the total period, expressed in days, over which the accumulated exposure to a contaminant is averaged. For non-carcinogenic risk, the value most frequently reported in the studies was 10,950 days for adults [86,90,91,92,93] and 2190 days for children [29,33,94,95,96]. In addition to these values, 25,550 days for adults was also often encountered [3,34,42,65,97]. However, some authors used values of 8760 days [31,98,99], 10,500 days [58], and 18,250 days [100]. When children are considered, instead of the averaging time value of 2190 days, other values such as 3650 days [43,67,76,77], 4380 days [101], 2100 days [58,78], and 1460 days [40,68] were sporadically used. Regarding carcinogenic risk, authors were largely consistent in using a value of 25,550 days for adults [102,103,104,105,106], which accounted for 70% of the screened studies assessing carcinogenic risk. The next most common value was 10,950 days [50,70,81,107,108], with a share of 13.8%. However, in some literature, averaging time values for adults and carcinogenic risk of 14,600 days [67], 21,900 days [35], 23,360 days [55], 24,930 days [44], and 26,280 days [109] were also used. In the case of children, 40.3% of studies on carcinogenic risk used a value of 25,550 days [27,98,110,111], while 36.4% used a value of 2190 days [46,88,112,113]. In addition to these values, some studies reported values of 3650 days [72,76], 4380 days [36], 2100 days [78], 3285 days [114], 26,280 days [109], and 23,725 days [28]. Although different averaging time values have been used across studies, for non-carcinogenic effects, averaging time is usually equal to the exposure duration (in days). However, for carcinogenic effects, averaging time is typically set to a lifetime (in days), because this risk is assumed to accumulate over an entire lifetime.

3.6. Skin Surface Area

Skin surface area is the area of the skin that is in contact with water, whose impact on health is being examined, and through which dermal contact occurs. It is expressed in cm². When it comes to adults, in the vast majority of the studies examining dermal health risk the value of 18,000 cm² [21,28,79,93,95,115,116] was used, accounting for 75.5%. Sporadically, in addition to this value, there appeared values such as 19,800 cm² [109], 17,000 cm² [114], 16,600 cm² [39,49], as well as 5700 cm² [31]. Children, since they are smaller and less developed, are characterized by lower skin surface area values. Here too the authors were quite in agreement, so that in 71.4% of the papers the value of 6600 cm² [26,54,61,107] was used for children. Still, some authors for skin surface area for children employed values of 6365 cm² [115], 6660 cm² [98], as well as 12,000 cm² [39,49]. Interestingly, a large number of studies (71.4%) used the combination of 18,000 cm² and 6600 cm² for adults and children, respectively. Similarly to the variability observed for body weight among the studies, skin surface area values also differed, which is expected given that skin size is directly related to overall body size and weight.

3.7. Exposure Time

Exposure time (h/day) represents the average daily duration of skin contact with water, reflecting how long an individual is exposed through dermal pathways each day. Regarding this parameter, the literature shows a notable consistency, with the majority of studies (69.6% and 67.4%) using the value of 0.58 h/day for adults and 1 h/day for children [28,29,42,50,79,95,111]. This was also the combination that appeared dominantly in the studies (65.2%). In addition to these values, other combinations encountered included 0.25 h/day and 0.25 h/day [31], 0.33 h/day and 0.33 h/day [39], 0.4 h/day and 0.4 h/day [101], 0.6 h/day and 0.6 h/day [34,98], and 1 h/day and 0.58 h/day [52,74,117] for adults and children, respectively.

3.8. Conversion Factor

The conversion factor (L/cm³) is used to convert the volume of water in contact with the skin from liters to cubic centimeters, ensuring unit consistency in the calculation of dermal chronic daily intake. In the literature, the authors were largely in agreement, so 91.5% of studies used a value of 0.001 L/cm³ [5,43,66,104,116]. However, some authors used values such as 0.000001 L/cm³ [118], 0.002 L/cm³ [88], and 0.01 L/cm³ [53] in their calculations. Since this parameter is used to convert liters to cm³, its value must be 0.001 L/cm³. Furthermore, this parameter should be included only in the calculation of chronic daily intake for the dermal route and should not be applied for the ingestion route. Specifically, if this parameter were excluded for the dermal risk assessment, chronic daily intake would have units of mg × cm³/(kg × day × L) and could not be used for calculating the hazard index and cancer risk, as the resulting values would not be dimensionless. This is because reference doses and cancer slope factors have units of mg/(kg × day) and (kg × day)/mg, respectively.

3.9. Dermal Permeability Coefficient

Dermal permeability coefficient represents the rate at which a chemical penetrates the skin barrier, expressed in cm/h. With regard to the values of dermal permeability coefficients, authors were largely consistent, and in agreement with those recommended by the USEPA. The value of 1 × 10⁻³ cm/h, recommended by USEPA for Cd, Hg, Cu, Mn, Fe, and As, was most commonly applied in the reviewed studies [29,33,95]. Similarly, values of 2 × 10⁻³ cm/h, 1 × 10⁻⁴ cm/h, 6 × 10⁻⁴ cm/h, 2 × 10⁻⁴ cm/h, and 4 × 10⁻⁴ cm/h were predominantly used for Cr(VI), Pb, Zn, Ni, and Co [21,59,96,117], respectively, also in accordance with the USEPA recommendations. For the remaining HMs, including Mo, Al, Sb, V, Se, Sr, Ba, and Sn, the available literature did not provide sufficient data to draw reliable conclusions, as these elements were rarely addressed in water studies. However, for these metals, the USEPA recommends a default value of 1 × 10⁻³ cm/h.

3.10. Reference Dose

Reference dose represents an estimate of the daily exposure to a chemical substance that is likely to be without appreciable risk of adverse health effects over a lifetime. Reference dose is expressed in mg/(kg × day) and has distinct values for different exposure routes to toxic substances from water, i.e., separate reference doses are determined for dermal exposure and for ingestion. A lower reference dose indicates a higher risk of non-carcinogenic health effects associated with the intake of heavy metals into the human body.

For Cd, authors predominantly use a reference dose value of 5 × 10⁻⁴ mg/(kg × day) for the ingestion route [29,33,67,84,88,95]. However, the use of 1 × 10⁻³ mg/(kg × day) is also reported, at roughly half the frequency [45,46,58]. Sporadically, in addition to these values, the following ingestion reference doses appear in the literature: 5 × 10⁻⁵ mg/(kg × day) [20], 1 × 10⁻⁴ mg/(kg × day) [26], 3 × 10⁻⁴ mg/(kg × day) [79], 3.5 × 10⁻³ mg/(kg × day) [113], and 0.5 mg/(kg × day) [66]. In the context of dermal exposure, the value of 2.5 × 10⁻⁵ mg/(kg × day) is most frequently reported [53,62,114], followed by 5 × 10⁻⁶ mg/(kg × day) [39,49,52] and 1 × 10⁻⁵ mg/(kg × day) [93,96,116]. Less commonly, values such as 2 × 10⁻⁵ mg/(kg × day) [79], 5 × 10⁻⁴ mg/(kg × day) [88], 1 × 10⁻³ mg/(kg × day) [46], and 5 × 10⁻³ mg/(kg × day) [78] are also cited in the literature.

With respect to Cr(VI) intake via water ingestion, the vast majority of studies have used a reference dose of 3 × 10⁻³ mg/(kg × day) [57,86,109]. In addition to this value, authors have also employed 1.5 mg/(kg × day) [44,97], 0.11 mg/(kg × day) [47], and 9 × 10⁻⁴ mg/(kg × day) [56]. For dermal risk, the reference dose for Cr of 7.5 × 10⁻⁵ mg/(kg × day) [28,53,79] has been predominantly used in the literature. The next most frequently reported values for dermal reference dose are 6 × 10⁻⁵ mg/(kg × day) [5,98] and 1.5 × 10⁻⁵ mg/(kg × day) [49,95]. Occasionally, values such as 3 × 10⁻³ mg/(kg × day) [88], 1.5 × 10⁻² mg/(kg × day) [78], 2.5 × 10⁻² mg/(kg × day) [50], and 1.5 mg/(kg × day) [115] are also reported.

In the case of Hg, authors were largely in agreement. The most frequently used ingestion reference dose was 3 × 10⁻⁴ mg/(kg × day) [21,33,104], while the most commonly applied dermal reference dose was 2.1 × 10⁻⁵ mg/(kg × day) [30,52,98]. In addition, some studies have reported the use of 1 × 10⁻⁴ mg/(kg × day) [95] and 1.6 × 10⁻⁴ mg/(kg × day) [109] for the ingestion route, as well as 1.3 × 10⁻³ mg/(kg × day) [88] and 3 × 10⁻³ mg/(kg × day) [5] for the dermal route.

For Pb, the ingestion reference dose most widely adopted by authors was 3.5 × 10⁻³ mg/(kg × day) [45,62,95], followed by 1.4 × 10⁻³ mg/(kg × day) [32,74] and 3.6 × 10⁻³ mg/(kg × day) [31,82], respectively. In addition to these, some authors have used values such as 4 × 10⁻³ mg/(kg × day) [24,71], 3.6 × 10⁻² mg/(kg × day) [27,85], and 4 × 10⁻² mg/(kg × day) [58,113]. In the calculation of dermal risk from Pb in water, the reference dose value of 4.2 × 10⁻⁴ mg/(kg × day) [54,60,81] has been predominantly used. However, values such as 5.25 × 10⁻⁴ mg/(kg × day) [28,93,98] and 3.5 × 10⁻³ mg/(kg × day) [109] have also been reported sporadically.

For Zn, only two ingestion reference dose values were encountered in the reviewed studies, with the most frequently used being 0.3 mg/(kg × day) [23,117,119], and a single occurrence of 2.4 × 10⁻² mg/(kg × day) [6]. In contrast, there was greater variability in the reported dermal reference dose values. The most commonly applied value was 6 × 10⁻² mg/(kg × day) [59,77,120], although other reported values included 6 × 10⁻³ mg/(kg × day) [39], 0.2 mg/(kg × day) [50], 0.3 mg/(kg × day) [109], 0.6 mg/(kg × day) [89], and 60 mg/(kg × day) [78].

With regard to Cu, the ingestion reference dose most frequently reported in the literature was 4 × 10⁻² mg/(kg × day) [54,62,87]. However, some studies have used alternative values such as 5 × 10⁻³ mg/(kg × day) [22,82], 3.7 × 10⁻² mg/(kg × day) [72,115], 1 × 10⁻³ mg/(kg × day) [113], and 7 × 10⁻⁴ mg/(kg × day) [6]. For the dermal exposure pathway, authors have predominantly employed 1.2 × 10⁻² mg/(kg × day) [77,89,98], although 8 × 10⁻³ mg/(kg × day) [20,114] has also been reported, with approximately three times lower frequency of occurrence. Occasionally, dermal reference dose value of 4 × 10⁻² mg/(kg × day) [46,78,81] has also been utilized in the literature.

In the case of Ni authors were generally in agreement. The ingestion reference dose most frequently used was 2 × 10⁻² mg/(kg × day) [33,45,58], while 5.4 × 10⁻³ mg/(kg × day) [81,89,110] was the predominant value for the dermal reference dose. In some studies, ingestion reference doses of 2 × 10⁻³ mg/(kg × day) [6] and 3 × 10⁻³ mg/(kg × day) [65] were also reported, whereas for the dermal route, additional values such as 8 × 10⁻⁴ mg/(kg × day) [21,107], 3 × 10⁻⁴ mg/(kg × day) [26], 4 × 10⁻² mg/(kg × day) [50], and 5.4 mg/(kg × day) [78] were encountered.

For Mn, the most frequently reported values in the literature are an ingestion reference dose of 0.14 mg/(kg × day) [26,44,119] and a dermal reference dose of 9.6 × 10⁻⁴ mg/(kg × day) [20,31,52]. An ingestion reference dose of 2.4 × 10⁻² mg/(kg × day) [29,45,107] is reported in roughly half as many studies, while the dermal reference dose of 1.84 × 10⁻³ mg/(kg × day) [5,104,116] appears with approximately 40% lower frequency. In addition, some authors have used values of 2 × 10⁻² mg/(kg × day) [69,120] or 1.4 × 10⁻² mg/(kg × day) [28,112] for the ingestion exposure route, as well as 8 × 10⁻⁴ mg/(kg × day) [74,77] for the dermal exposure route.

For Fe, the ingestion reference dose most frequently reported in the literature was 0.7 mg/(kg × day) [3,58,72], followed by 0.3 mg/(kg × day) [44,81,88], which appeared with approximately 60% lower frequency. Additionally, values of 7 × 10⁻³ mg/(kg × day) [78,119] and 3.1 × 10⁻² mg/(kg × day) [6] were reported sporadically. For the dermal exposure route, the most prevalent value was 0.14 mg/(kg × day) [53,54,114], with 4.5 × 10⁻² mg/(kg × day) [34,49] occurring about 40% less frequently. Nonetheless, values of 0.2 mg/(kg × day) [50], 0.3 mg/(kg × day) [101], and 0.7 mg/(kg × day) [46] have also been reported in the literature as dermal reference doses.

With regard to As, a high level of agreement was observed among authors in the use of the ingestion reference dose. Specifically, the vast majority of studies applied a value of 3 × 10⁻⁴ mg/(kg × day) [45,95,99], while the remaining four studies reported values of 3 × 10⁻⁵ mg/(kg × day) [114], 1.23 × 10⁻⁴ mg/(kg × day) [79], 4 × 10⁻⁴ mg/(kg × day) [24], and 3 × 10⁻³ mg/(kg × day) [92]. For the dermal exposure route, the level of agreement was lower, with the most frequently reported value being 1.23 × 10⁻⁴ mg/(kg × day) [30,93,116], followed by 2.85 × 10⁻⁴ mg/(kg × day) [107,109,114], which occurred approximately 40% less often. In addition, values of 1 × 10⁻⁴ mg/(kg × day) [88] and 1.2 × 10⁻⁴ mg/(kg × day) [52] were also reported for the dermal route.

For HMs that have been rarely investigated in the literature, the order of occurrence in the assessment of ingestion-related non-carcinogenic risk, expressed as a percentage of the total number of analyzed studies, was as follows: Co (21.7%) > Al, Ba (13.4%) > Mo (10.3%) > Sr (9.3%) > V, Sb (7.2%) > Se (6.2%) > Sn (2.1%). In contrast, dermal risk associated with these HMs has been scarcely examined, with an average of only about three studies addressing it. For Co, the ingestion reference dose most commonly reported was 3 × 10⁻⁴ mg/(kg × day) [45,56,58], followed, at roughly half the frequency, by 2 × 10⁻² mg/(kg × day) [31,87,100]. For Al, Ba, Mo, and Se authors were fairly consistent, most often applying ingestion reference doses of 1 mg/(kg × day) [30,57,97], 0.2 mg/(kg × day) [6,51,92], 5 × 10⁻³ mg/(kg × day), and 5 × 10⁻³ mg/(kg × day) [27,32,86,108], respectively. In the case of Sr and Sn, only a single ingestion reference dose value of 0.6 mg/(kg × day) was reported in the literature for each [22,49,56,118]. Similarly, a single ingestion reference dose value was reported for Sb, namely 4 × 10⁻⁴ mg/(kg × day) [27,92]. For V, however, the situation was somewhat different, as several ingestion reference dose values were noted, including 5.4 × 10⁻³ mg/(kg × day) [116], 9 × 10⁻³ mg/(kg × day) [65], 5 × 10⁻³ mg/(kg × day) [45], and 7 × 10⁻³ mg/(kg × day) [72]. Regarding dermal exposure to these HMs, dermal reference dose values were reported in only a single study for Mo (1.9 × 10⁻³ mg/(kg × day) [120]), Sr (0.12 mg/(kg × day) [49]), and V (1.0 × 10⁻⁵ mg/(kg × day) [120]), as well as in two studies for Ba (0.2 and 14 mg/(kg × day) [78,92]). Three studies examined the dermal non-carcinogenic risk of Sb, applying reference doses of 8.0 × 10⁻⁶ [117], 8.4 × 10⁻⁶ [116], and 4.0 × 10⁻⁴ mg/(kg × day) [92]. Finally, for Co, eight studies reported dermal reference doses, most frequently 1.6 × 10⁻² mg/(kg × day) [59,74], followed by 6.0 × 10⁻⁵ mg/(kg × day) [81] and 5.4 × 10⁻³ mg/(kg × day) [31].

These findings highlight the considerable gaps and inconsistencies in the available data for these rarely investigated HMs. In particular, the very limited number of studies addressing dermal exposure (Table S2) raises uncertainty regarding the reliability and applicability of risk estimates. Moreover, the other HMs also exhibited considerable variation, and dermal risk for these HMs has also been investigated much less frequently.

3.11. Cancer Slope Factor

The cancer slope factor is a parameter used to estimate the lifetime cancer risk from chronic exposure to a chemical, representing the increased probability of cancer occurrence per unit of daily intake, expressed in (kg × day)/mg. A higher value of this parameter indicates that the given HM carries a greater risk of cancer development. The HMs with carcinogenic effects on human health that have been investigated in the literature include As, Cd, Cr(VI), Pb, and Ni, including both ingestion and dermal routes.

For Cd, the ingestion slope factor most frequently reported by authors was 6.1 (kg × day)/mg [24,33,50], although the value of 0.38 (kg × day)/mg [26,44,95] also appeared relatively often. Together, these two values accounted for approximately three-quarters of all studies. Nevertheless, some studies also reported alternative values such as 0.61 (kg × day)/mg [54,56,84], 6.3 (kg × day)/mg [61,89], and 15 (kg × day)/mg [31,77,107]. With regard to dermal risk, the same two values that were predominantly used for the ingestion route were also reported, with 6.1 (kg × day)/mg slightly prevailing in this case as well. In addition, values of 0.5 (kg × day)/mg [49], 1.5 (kg × day)/mg [92], and 6100 (kg × day)/mg [50] were also reported.

When it comes to Cr(VI), the ingestion slope factor most commonly identified in the literature was 0.5 (kg × day)/mg [5,42,108], although values such as 0.05 (kg × day)/mg [44], 0.041 (kg × day)/mg [121], and 41 (kg × day)/mg [62,117] were also occasionally reported. In contrast, for the dermal slope factor, two values predominated: 20 (kg × day)/mg [30,33,98], which appeared slightly more frequently, and 0.5 (kg × day)/mg [26,52,114], which followed. This indicates a lack of consensus among authors regarding the dermal slope factor for Cr(VI). In addition to these two values, 0.132 (kg × day)/mg [74] and 0.2 (kg × day)/mg [89] were also reported in certain studies.

For Pb, authors generally agreed on a single ingestion slope factor value, with 8.5 × 10⁻³ (kg × day)/mg prevailing in approximately 77% of cases [28,95,96]. In addition, some studies employed alternative values such as 4.2 × 10⁻² (kg × day)/mg [119], 8.5 × 10⁻² (kg × day)/mg [108], and 0.5 (kg × day)/mg [50]. Carcinogenic risk via the dermal pathway for Pb has not been widely assessed in the literature. However, a few authors have calculated it using dermal slope factors of 8.5 × 10⁻³ (kg × day)/mg [29], 7.3 × 10⁻² (kg × day)/mg [33,117], and, in some cases, 8.5 (kg × day)/mg [78,81], and 7.3 (kg × day)/mg [74] and 500 (kg × day)/mg [50].

When it comes to the cancer risk associated with Ni, the literature is very limited. Ingestion slope factor values for Ni were reported in only about 14% of the total studies, while dermal slope factor values appeared in an even smaller proportion, around 5%. Within this limited body of work, there was no consensus among authors: the ingestion slope factor was most often reported as 0.84 (kg × day)/mg [22,89,119] and 1.7 (kg × day)/mg [1,100]. For the dermal route, a wide range of values has been used, including 42.5 (kg × day)/mg [33,118], 2 × 10⁻² (kg × day)/mg [26], 0.64 (kg × day)/mg [74], and 0.84 (kg × day)/mg [78]. These inconsistencies raise questions about the reliability and relevance of cancer risk assessment for Ni in water.

For As, authors were highly consistent. In more than 85% of studies, values of 1.5 (kg × day)/mg and 3.66 (kg × day)/mg were reported for ingestion and dermal slope factors, respectively [62,95,99,107]. Nevertheless, sporadic use of alternative values was observed, including 3 × 10⁻⁴ (kg × day)/mg [40], 0.5 (kg × day)/mg [113], and 15 (kg × day)/mg [117,119] for the ingestion route, as well as 1.5 (kg × day)/mg [101] and 1.58 (kg × day)/mg [109] for the dermal route.

4. Probabilistic Health Risk Assessment

Health risk assessment of HMs in water has predominantly been conducted using a deterministic approach, which takes into account single values of exposure parameters. This way of assessment is characterized by speed, less computational time, and simplicity. However, such an approach may lead to health risks being either overestimated or underestimated. The reason for this lies in the uncertainty and variability of exposure parameters. Uncertainty refers to the lack of precise knowledge about the true values of parameters used in the risk assessment, while variability refers to the inherent differences among individuals or populations in their responses to a hazard. Uncertainty can be reduced, whereas variability cannot be reduced but only better characterized [18].

This issue can be addressed by introducing a probabilistic approach to risk assessment, which is performed through a Monte Carlo simulation. A Monte Carlo simulation, instead of single values of exposure parameters, uses appropriate probabilistic distributions, thereby allowing a whole spectrum of risk values to be obtained, with the possibility to state with what probability a certain risk value can be expected. Although it requires more knowledge and time as well as computer time, it has been shown to provide more precise and accurate risk estimates [2,30].

A Monte Carlo simulation is conducted in several steps: (1) defining exposure parameters that are characterized by uncertainty/variability and assigning distributions, (2) defining output parameters, i.e., risk estimates, (3) setting working parameters, such as sampling method, number of trials, and confidence level, and (4) running the simulation [104,122]. Monte Carlo simulation is performed through repeated random sampling of probabilistic distributions, with thousands of iterations (commonly 10,000) and a 95% confidence level, to generate reliable estimates of health risks.

Among the studies reviewed, 54 studies applied probabilistic risk assessment, representing only 25.7%. However, the majority of these studies did not specify which distributions were assigned to the input parameters of the assessment. Probabilistic distributions were reported in only 19 studies. Regarding HM concentrations, authors were consistent and used lognormal distributions [30,35,100]. For ingestion rate, the vast majority of authors adopted normal distribution [25,123], while some used lognormal [32] or triangular [120]. In the case of exposure frequency, authors were also in agreement, as triangular distribution [60,117,124] was found in the literature, although one study applied a point distribution [109]. With respect to averaging time, exposure time, and conversion factor, authors were consistent and employed point distributions in all three cases [35,52,96]. For body weight, most studies used a lognormal distribution [99,125,126], but about 35% fewer studies employed a normal distribution [26,109]. For skin surface area, the majority of works employed a lognormal distribution [110,116,124]. However, some authors also applied point [52], normal [109], and triangular [120] distributions. For exposure time, authors used either point [117,125] or lognormal [30,96] distributions, with the latter occurring about 30% less frequently. However, it would be expected that this parameter should not be represented by a point value, but rather exhibit variability across the population. Finally, with respect to dermal permeability coefficients, reference doses and cancer slope factors, these were always treated as point values.

5. Summary Tables

Through the review of studies applying deterministic and probabilistic approaches to health risk assessment in water, the most commonly adopted values of exposure parameters and probability distributions were identified and are summarized in the tables below. The selected values, verified and compared with those recommended by the USEPA, reflect the parameters most often used by researchers and may serve as practical recommendations for future use.

Table 1. Exposure factors used in the calculation of chronic daily intake for health risk assessment in water.

Parameter	Abbreviation	Unit	Population Group		Probability Distribution
			Adults	Children
Heavy metal(loid) concentration	C	mg/L	/	/	Lognormal
Ingestion rate	IR	L/day	2	1	Normal
Exposure frequency	EF	days/year	365	365	Triangular
Exposure duration	ED	years	30	6	Point
Body weight	BW	kg	70	15	Lognormal
Averaging time	AT	days	10,950 nc	2190 nc	Point
			25,550 c	25,550 c
Skin surface area	SA	cm2	18,000	6600	Lognormal
Exposure time	ET	h/day	0.58	1	Lognormal
Conversion factor	CF	L/cm3	0.001		Point

Notes: nc: non-carcinogenic risk; c: carcinogenic risk.

presents the exposure parameters included in the calculation of chronic daily intake, including their units, values for adults and children, as well as the probability distributions applied in the probabilistic risk assessment model.

Table 2. Dermal permeability coefficients (Kp) and reference doses (RfD) of heavy metal(loid)s employed in the health risk assessment in water.

Heavy Metal(loid)	Dermal Permeability Coefficient (Kp), cm/h	Reference Dose (RfD), mg/(kg × day)		Probability Distribution
		Ingestion Route	Dermal Route
Cd	1.0 × 10−3	5.0 × 10−4	2.5 × 10−5	Point
Cr(VI)	2.0 × 10−3	3.0 × 10−3	7.5 × 10−5
Hg	1.0 × 10−3	3.0 × 10−4	2.1 × 10−5
Pb	1.0 × 10−4	3.5 × 10−3	4.2 × 10−4
Al	1.0 × 10−3	1.0 × 100	-
Zn	6.0 × 10−4	3.0 × 10−1	6.0 × 10−2
Cu	1.0 × 10−3	4.0 × 10−2	1.2 × 10−2
Ni	2.0 × 10−4	2.0 × 10−2	5.4 × 10−3
Sb	1.0 × 10−3	4.0 × 10−4	-
V	1.0 × 10−3	5.04 × 10−3	-
Mn	1.0 × 10−3	1.40 × 10−1	9.6 × 10−4
Fe	1.0 × 10−3	7.0 × 10−1	1.4 × 10−1
As	1.0 × 10−3	3.0 × 10−4	1.23 × 10−4
Co	4.0 × 10−4	3.0 × 10−4	1.60 × 10−2
Se	1.0 × 10−3	5.0 × 10−3	-
Sr	1.0 × 10−3	6.0 × 10−1	-
Ba	1.0 × 10−3	2.0 × 10−1	-
Sn	1.0 × 10−3	6.0 × 10−1	-
Mo	1.0 × 10−3	5.0 × 10−3	-

and

Table 3. Cancer slope factors (SFs) of heavy metal(loid)s employed in the health risk assessment in water.

Heavy Metal(loid)	Cancer Slope Factor (SF), (kg × day)/mg		Probability Distribution
	Ingestion Route	Dermal Route
Cr(VI)	0.5	20	Point
As	1.5	3.66
Cd	6.1	6.1
Pb	8.5 × 10−3	-

summarize the dermal permeability coefficients, reference doses and cancer slope factors for HMs, also including their units and probability distributions.

With regard to the exposure factors presented in Table 1, the values most frequently reported across the reviewed studies largely coincide with those recommended by the USEPA, namely exposure duration (30 years for adults and 6 years for children), exposure time (0.58 h/day for adults and 1 h/day for children), averaging time (non-carcinogenic risk: 10,950 days for adults and 2190 days for children; carcinogenic risk: 25,550 days for both adults and children), body weight (70 kg for adults and 15 kg for children), and skin surface area (18,000 cm² for adults and 6600 cm² for children). However, for exposure frequency, the value most commonly used in the studies was 365 days, whereas USEPA recommends 350 days, assuming that individuals spend approximately two weeks per year away from their primary residence. Regarding ingestion rate, the USEPA provides different values for various age groups rather than a single value for adults and another for children. Nevertheless, earlier versions of the USEPA health risk assessment guidelines reported average values of 2 L/day for adults and 1 L/day for children, which were most commonly adopted in the reviewed studies.

When it comes to reference dose values, HMs that appeared in less than 5% of the reviewed studies were not included in the summary Table 2. In addition, HMs were also excluded when, although occurring in more than 5% of studies, no single reference dose value was reported in at least 5% of cases, but instead the values were scattered across studies. Such was the case with Sn, which was addressed in only 2.1% of studies. In this instance, the value proposed by the USEPA was reported. However, it is important to note that the USEPA value coincided with those found in the literature. Similarly, the same situation applied to dermal reference doses for Al, Sb, Se, V, Sr, Ba, and Mo, where the number of available studies was insufficient to draw reliable conclusions. Additionally, the USEPA [18] has not established dermal reference doses or cancer slope factors for chemicals. Therefore, dermal reference doses for these HMs were omitted from Table 2. In addition, the dermal cancer slope factor for Pb was excluded from Table 3 due to the lack of consensus among authors regarding a specific value.

It is important to note that the ingestion reference dose values most frequently reported in the studies were those recommended by the USEPA, particularly for HMs such as Cd, Zn, Cu, Ni, Mn, Mo, Hg, Sn, Ba, Sr, Se, Co, Sb, Al, and Fe. In the case of Pb, Cr(VI), and As, Table 2 presents the values most commonly used in the reviewed studies, as this paper aims to reflect the practices reported in the literature. On the other hand, for V, the available studies were insufficient, and therefore the ingestion reference dose was taken from the USEPA database. Table 2 also presents the dermal permeability coefficients, showing the most commonly applied values for Cd, Hg, Cu, Mn, Fe, Cr(VI), Pb, Zn, Ni, Co, and As, which were consistent with the USEPA recommendations. For Mo, Al, Sb, V, Se, Sr, Ba, and Sn, due to the limited number of studies, the default USEPA value of 1 × 10⁻³ cm/h was applied. When it comes to cancer slope factors, the reviewed literature and USEPA values differed completely. Therefore, Table 3 presents those values reported in the studies, as this paper aims to reflect the practices documented in the literature.

Overall, these data highlights significant variability in exposure and toxicity factor choices across studies. We recommend harmonizing key parameters to the values summarized in Table 1, Table 2 and Table 3 for consistency in future risk assessments. Adopting these commonly used values will improve comparability of HRA results and reduce uncertainty. Likewise, researchers should transparently justify any deviations from these standard factors.

6. Conclusions and Future Perspectives

In this review, an overview of exposure and toxicity parameters used in health risk assessment of HMs in water was conducted. In this regard, the analysis included exposure factors such as ingestion rate (IR), exposure frequency (EF), exposure duration (ED), body weight (BW), averaging time (AT), skin surface area (SA), exposure time (ET), dermal permeability coefficient (Kp), and conversion factor (CF), as well as toxicity factors, namely reference dose (RfD) and cancer slope factor (SF).

The occurrence of individual parameter values was discussed, along with their comparison with the USEPA recommendations. Finally, summary tables were compiled, including the most frequently applied values of the mentioned parameters, which may serve as a reference or recommendation for future researchers in selecting parameters for health risk assessment of HMs in water.

The results indicate that for the majority of parameters, the values were the same as those prescribed by the USEPA. However, it is important to note that some parameters deviated from this practice, namely ingestion rate and exposure frequency. Moreover, certain HMs, including Mo, Al, Sb, V, Se, Sr, Ba, and Sn, have not been sufficiently studied in the literature, and therefore, reliable conclusions regarding the most frequently applied dermal permeability coefficients, reference doses, and cancer slope factors could not be drawn. This highlights the need for future studies to investigate these elements in water in order to cover the full spectrum of HMs in risk assessments.

Additionally, this review concluded that authors primarily focused on the ingestion exposure route, while very often neglecting the dermal exposure route. Furthermore, the vast majority of studies employed a deterministic approach to risk assessment, while only a small number applied a probabilistic approach. Finally, a clear geographical research gap was identified, as the majority of studies were concentrated in Southeast Asia, with limited coverage in Africa and only sporadic work in Europe, while North America and Australia were absent.

In summary, in future studies it is desirable to include both exposure routes of HMs in water, as well as to apply probabilistic risk assessment, which has been shown to provide more precise and accurate outcomes. There is also a clear need to extend health risk assessment of HMs in water to regions that remain underrepresented.